Content-Type: text/html

Free Riders and Pretenders:
 Media Industry Organizations and Collective Action





Jennifer H. Meadows
Associate Professor
Department of Communication Design
California State University, Chico
Chico, CA  95929
530/898-4775
Email:  <a href="/cgi-bin/wa?LOGON=A3%3Dind0209A%26L%3DAEJMC%26E%3D8bit%26P%3D3176333%26B%3D--%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D_24457960%253D%253D_%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>

August E. Grant
Focus25 Research & Consulting
25 Crossbow Lakes Court
Columbia, SC  29212
803/749-3578
Email:  <a href="/cgi-bin/wa?LOGON=A3%3Dind0209A%26L%3DAEJMC%26E%3D8bit%26P%3D3176333%26B%3D--%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D%253D_24457960%253D%253D_%26T%3Dtext%252Fplain%3B%2520charset%3Diso-8859-1" >[log in to unmask]</a>



Paper Submitted to the
Media Management and Economics Division
AEJMC
March 30, 2002

Running head:  Free Riders and Pretenders








Free Riders and Pretenders: Media Industry Organizations and Collective Action:

Abstract


This study uses collective action theory to examine membership in media industry
organizations. In particular it examines the free riders, those actors who benefit from the
services of a media industry organization such as lobbying but do not participate.  A
survey of members and non-members of a media organization was conducted to measure
dimensions of collective action. A third group, pretenders, non-members who indicate
that they are members, was found.  Differences among members, non-members and
pretenders are examined.


Free Riders and Pretenders:  Media Industry Organizations and Collective Action

Introduction

Various media industry organizations work to help the good of the industry as well as
their members.  For example, state broadcasting organizations serve a number of
functions for radio and television stations.  Lobbying, training, and information sharing
are just a few of the services provided by these organizations.  While membership dues
finance these organizations, many of the services performed by state broadcasting
organizations affect the media industry as a whole thus benefiting both members and
non-members alike.  For example, lobbying the state legislature benefits all of the stateŐs
broadcasters, not just members.  Thus, one problem that state broadcasting organizations
face, as well as other media industry organizations, is membership, as a percentage of
broadcasters do not become members because they can reap many of the benefits of the
organization without the cost of joining.
In this study, we have chosen to use collective action theory to examine reasons
why stations choose to join or not join their state broadcasting organization.  Collective
action theory explains and explores the reasons why individuals choose whether or not to
participate in groups working for the collective good.   In this case, collective action
theory will help us understand why stations choose to "free ride," meaning enjoy the
benefits of state broadcasting organizationŐs work without becoming a member.



Collective Action Theory

Collective action theory is concerned with the manner in which people work
together to reach a common goal.  Much of the work in collective action theory has been
concerned with collective action problems.  These problems arise when individuals in a
group have incentives, which are contrary to the common goal. Marwell and Oliver
(1993) write: "the problems of collective action is thus the conflict between individual
group memberŐs pursuit of their own interest and the needs of the group as a whole." (p.
5).   This self-interested behavior then leads to collective action problems such as free
riding.  Goetze (1994) writes of collective action: "...cooperative actions by individuals
generate benefits for everyone in the group, but the self-interested individual benefits
even more from choosing not to cooperate.... In other words, the contemporary sense of
collective action refers to situations in which cooperation is desirable but problematic."
(p. 56).
        Collective action theory was created to explain behaviors that appear to some to
be "non-rational", i.e., actions by individuals that seem counter to the economic self
interest of the individual (Marwell & Oliver, 1993).  Through the work of Mancur Olson
(1965) and others (Heckathorn, 1993; Taylor & Singleton, 1993; Goetze, 1994)),
sociologists and economists have developed a theory that illustrates the manner in which
these "altruistic" behaviors, which provide more benefit to the group than to the
individual, can be seen as contributing to the self-interest of the individual.  There will,
however, be "rational" behaviors such as free riding.  Heckathorn (1993) suggests that
there are three conditions for free riding to occur.  First, the collective good is jointly
produced.  Second, the production of the collective good gives benefits to all people in
the group. Finally, the production of the collective good includes costs.
Olson (1965) writes in his seminal work, The Logic of Collective Action, that
there is a dilemma connected to collective action, one in which a "rational" actor will try
to free-ride, or collect the benefits of others actions without participating themselves.
Free riders serve their own self-interest by not participating because they know they will
obtain the benefits without the cost of participation.  Olson also explains that actors may
free-ride if they believe that their contribution will not have a positive effect on the
outcome -- in other words, their actions wonŐt make a difference, so why participate?
        Olson also argues that collective action is a function of group size.  Small groups,
according to Olson, do not have a free-rider problem because each individualŐs
contribution is so important that there are real incentives, including social incentives such
as friendship and social status, to participate.  In addition, in a small group, one single
member may receive so much benefit as to supply the cost for the rest of the group.  In
intermediate groups, no one individual benefits so greatly as to take over the cost,
however, the group is still small enough to not allow the free riders to hide.  Thus, there
is a greater chance for collective action than not.  However, with a large group,
"individual benefit is definitely too low and organization costs too high, for collective
action to be possible," according to Olson (Udehn, 1996, p. 240).  Large groups, then,
must rely on incentives and coercion to make collective action work.
        OlsonŐs work on collective action was expanded upon and studied by other social
scientists.  For example, studies on collective action and group size have found that,
contrary to Olson, group size does not always have such a simple relationship to
collective action.  Larger groups do pose more of a problem, but they are not always a
problem (Marwell & Oliver, 1988).  In addition, incentives and coercion were disproven
as means to make collective action work (Marwell and Oliver 1993).
One major re-conceptualization of collective action theory put forth by Marwell
& Oliver (1993) suggests collective action can best be conceptualized as a production
function. The key to understanding collective action theory is identification of which of
three "production functions" is associated with the group activity.   In an accelerating
production function (Figure 1), the contributions of the first contributors make a
relatively small impact on the public good, with each successive contributor making a
greater contribution.  An example of an accelerating production function is the adoption
of interactive media such as the telephone or fax machine.  The first person to buy a
telephone or fax made almost no contribution to the public good, but each successive
contribution increased the value of the technology to all other potential adopters,
eventually resulting in universal adoption (Markus, 1987). In an accelerating production
function, the problem is not free riders but getting individuals or actors to "start-up" --
making the initial contribution.  As Heckathorn (1996) notes "Unless a critical mass of
strongly motivated individuals is willing to absorb these costs, collective action never
begins" (p. 251).

Figure 1:  Accelerating Production Function

In a linear production function (illustrated in Figure 2) the contributions of the
first contributors are equal to those of later contributors.  In the linear production
function, there are no free riders because the marginal gain is the same for early and late
contributors.  Everyone then participates because they will have the same gain.  On the
other hand, some individuals may not participate for these same reasons.

Figure 2:  Linear Production Function

The decelerating production function (Figure 3) is characterized by the fact that
the first contributors make a disproportionately large contribution to the public good,
with each successive contributor having a smaller impact.  Ultimately, the maximum
good for all members of a community is served well before all members of the group
have made a contribution.  The result is that the group members who have not made a
contribution get a "free ride" on the contributions of others.  Membership associations
that engage in lobbying and similar efforts that benefit all members of an industry, are
examples of a decelerating production function.

Figure 3: Decelerating Production Function

Marwell and Oliver (1993) argue that in this situation free riding comes out of
order effects and surplus.  With order effects, interested individuals contribute to the
common good while the less interested free ride on those contributions.  In addition, the
most interested contribute first while the least interested wait.  Ideally for full
participation, the least interested would contribute first as the rewards are the most
evident and then the most interested would contribute later when the rewards are less.
However, this is not what happens with order effects, thus leading to free riding.  With
surplus, free riding occurs as more individuals contribute to the common goal, allowing
other individuals to free ride or fail to contribute.  In this situation many actors are ready
to contribute in the beginning, but the collective goals can be met by the contributions of
a subset of those who are willing to make the first contributions.  Then, once these actors
realize that the collective goal has been achieved, many become unwilling to contribute
later.  Heckathorn (1996) notes: "This produces a surplus of contributors in the sense that
some of those who were initially willing to contribute will refuse if others have
contributed first" (p. 251).  At the most basic level, the difference between Olson and
Marwell and OliverŐs versions of free riders and collective action is that Olson believes
that free riding is an ever-present temptation while Marwell and Oliver believe that free
riding only occurs if there is a decelerating production function.
        Some interpretations put collective action theory at odds with classical,
rational models of economic behavior because the theory seems to call for "non-rational"
behavior on the part of the individualŃbehavior in which the individual acts on the
interest of others rather than his or her own self-interest.  In addition, collective action as
conceptualized by Oliver, Marwell & Teixeira (1985) assumes interdependent decisions
while the traditional economic assumption is that of isolated, independent decisions (For
a detailed discussion of the issue of rational vs. irrational self-interest, and interdependent
decisions see Marwell & Oliver, 1993.)  In addition Heckathorn (1993) notes that when a
person chooses whether or not to contribute there are actually two levels of choices.
First, the actor decides whether to participate and then whether to influence others to
participate.  Interdependence is very important.  Marwell and Oliver (1993) note that an
actorŐs choice varies depending on the choices of others.
Indeed, researchers examined the non-economic reasons to participate in
collective action such as social norms, altruism, and morality.  For collective action to
succeed there needs to be a belief in the efficacy of the group (Marwell and Oliver,
1993).  Interests and resources of individuals must be considered.  A number of
researchers have indicated that heterogeneity in resources and interests are important to
facilitate collective action.  This is especially important in the case of a decelerating
production function situation such as the TAB example.  There must be high
resource/high interest individuals to take up the fight for collective action.  The problem
is that, ideally, those individuals would join later in order to allow individuals with lower
resources and interest to see the value in their joining, thus, the example of order effects
Marwell and Oliver, 1993).
There are a number of approaches put forth to help solve free-riding problems,
including cooperation, selective incentives, and coercion.  Taylor and Singleton (1993)
argue that solving free riding or collective action problems is a three-part process.  First,
the group must "identify the possibilities for cooperation" (p. 196).  Second, the group
must agree on one "plan" for cooperation.  Finally, a method of monitoring or enforcing
the plan must be devised.  The authors note that often solving a collective action problem
is a costly procedure.   The first step involves search costs.  The second step involves
bargaining costs, and the final step involves monitoring and enforcement costs.  Because
solving collective action problems is so costly, the organization or group must decide if
100 percent participation is worth the cost.
In this article, we study the Texas Association of Broadcasters (TAB).  Meadows
and Grant (1997) studied TAB membership in 1994 and found collective action theory
useful in understanding the differences in attitudes regarding TAB between members and
non-members.  In a survey of Texas radio and television stations, the researchers ask
questions specifically related to collective action in order to see differences between
members and non-members.  The results indicate the utility of collective action theory in
understanding attitudes towards TAB membership, as well as larger issues related to
membership in trade organizations.

The Texas Association of Broadcasters

The Texas Association of Broadcasters offers a wide range of services that range
from those serving the entire industry such as lobbying to those serving individual
stations such as legal advice.  Monthly membership dues are collected from radio and
television stations based on yearly net billing. Other state broadcasting organizations use
revenues and other factors such as market size, rate card, and type of station to determine
dues (NAB, 1993).
As with other state broadcasting organizations, TAB is concerned with
membership growth and retention.  There is a paucity of research on state broadcasting
organizations.  Gold (1994) studied these organizations in Tennessee and North Carolina
and found that there were perceptual differences between small and large market stations.
For example, small market radio stations perceived that the state broadcasting
organization was focused more on the needs of large market stations than on small
market stations.  However, member stations believed that small market stations, radio
stations, and less affluent stations benefited more from member services than television,
affluent, and large market stations.
        Meadows and Grant (1997) took the next step and examined attitudes of members
and non-members of the Texas Association of Broadcasters within the framework of
collection action theory.  Their study found that members and non-members differ in
their ratings of the importance of employment (i.e. position listings), information (i.e.
newsletter), and legislative services (i.e. lobbying).  Member radio members ranked the
importance of membership services (i.e. legal advice, telephone discounts) higher than
member television stations.  In order to better understand differences between radio and
television stations, and members and non-members the researchers used collective action
theory.  Non-members were characterized as "free-riders."   Non-members did not care
about TABŐs work for "the good of the industry" but did not see the individual services
as valuable enough to join.  Thus, it was suggested that if TAB wanted to increase their
membership they may want to address membership services which affect individual
stations because non-members already knew the collective good was being taken care of
regardless of their participation and membership.

TAB and Collective Action

Collective Action theory suggests three factors that TAB should consider in order
to maximize membership.  First is the economic factor.  Non-members will free ride
because of economic self-interest.  If economic self-interest is not served by joining
TAB, what is the point in joining?  This attitude concurs with OlsonŐs (1965) idea of the
"rational" decision.  This economic self-interest could then be expected to be stronger
with radio stations than television stations.  Radio stations have smaller budgets in
general than television.  Thus, the action of joining TAB would be seen as more of a
hardship and not joining would be "rational" on these stations part.


The economic incentive to join suggests three hypotheses:

H1a.    Non-members will rate economic factors for joining TAB higher than members
H1b     Radio stations will rate economic factors higher than television stations

The second factor to consider is social norms.  In this case, we refer to this as
"peer pressure."  Stations may be joining TAB because other similar stations are
members also.  As noted collective action is an interdependent decision and one of the
factors in decision-making is what others are doing (Marwell & Oliver, 1993).

H2a.    Non-members will rate peer pressure higher than members.
H2b     Radio stations will rate peer pressure higher than television stations.

Altruism is another factor to consider.  Joining TAB may come out a genuine
wish to work for the good of the industry.  This could be described under collective
action theory as altruistic "non-rational" behavior.  The following hypotheses are
suggested.

H3a.    Members will rate welfare of the industry higher than non-members.
H3b     Radio stations will rate welfare of the industry lower than television stations.

        As discussed below, the definition of "membership" is more complicated than a
simple member/non-member dichotomy.  Self-report data on membership may conflict
with the records of the organization, with some actors who havenŐt paid dues claiming to
be members.  The reason for the difference between the self-report measure and the
objective measure (payment of dues) may be the result of a factor as simple as someone
forgetting to mail a check, or it may be the more complex product of psychological
factors related to self-comparison and self-image.
        Because this analysis is focused on underlying motivations, these actors who have
not paid dues but report to be members must be considered separately from members and
non-members.  The term "pretenders" is used for this classification of membership.  The
fact that pretenders have not paid dues suggests that they are "free riders," enjoying the
benefits of membership without paying the economic price of membership.  In this
respect, pretenders will be more similar to non-members than to members.  However,
because they are presenting themselves as being part of the membership organization, on
measures of peer pressure and welfare of the industry, pretenders will be more like
members.  Accordingly, we propose the following hypotheses:

H4a:    Pretenders will be more similar to non-members than to members on rating
importance of economic issues.
H4b:    Pretenders will be more similar to members than to non-members on rating
importance of peer pressure.
H4c:    Pretenders will be more similar to members than to members on rating importance
of the welfare of the industry.


Methods

The mail survey instrument was patterned after a similar instrument used in the
1994 TAB Membership Study to allow comparison of responses over time.  In addition to
evaluation of a variety of issues and services provided by TAB, the questionnaire
solicited descriptive information including station type (radio or television) and self-
report of membership status.  A five-item scale (Table 1) was created to measure the
different dimensions of collective action, with each item being rated on a five-point scale
from "Strongly Disagree" to "Strongly Agree."  Exploratory factor analysis (principal
components extraction with Varimax rotation) suggested a three-factor solution, with one
item each for the economic ("Joining TAB is a bottom-line decisionŃthe membership
must pay for itself") and peer pressure ("Stations similar to mine are TAB members")
dimensions.  Three items loaded on the third dimension, which included items related to
the collective good of the industry ("When a station joins TAB every broadcaster
benefits", "TAB makes a significant contribution to Texas broadcasters", and "There is
no need to join TAB because they already have enough members").  A scale for the
collective good was created by reversing the last item and summing the three items
(reliability, measured using coefficient alpha, was .75).

Table 1:
COLLECTIVE ACTION 5-ITEM SCALE

Strongly                         Strongly
  Disagree             Neutral           Agree
When a station joins TAB every broadcaster
benefits
1         2         3         4         5
Stations similar to mine are TAB members
1         2         3         4         5
There is no need to join TAB because they
already have enough members
1         2         3         4         5
TAB makes a significant contribution to
Texas broadcasters
1         2         3         4         5
Joining TAB is a bottom-line decisionŃthe
membership must pay for itself.
1         2         3         4         5

TAB provided a complete list of Texas radio and television managers.  Random
selection was used to generate a sample of 100 radio stations and 100 television stations.
This sample included a disproportionate number of members; it was therefore
supplemented by adding additional non-members.  Table 2 reports the sample
composition.

Table 2:
SAMPLE COMPOSITION

Radio
Television
Total
Member
49
63
112
Non-Member
52
49
101
Total
101
112
213

Two surveys were returned undeliverable, yielding a sample size of 211.  92
responses were received, yielding a response rate of 44%.  The following table reports the
responses by station type and membership.  In comparing the responses to the self-report
"membership" question with TAB records on membership, it became apparent that a
third category of membership was needed for non-members who indicated on their
surveys that they were members; as discussed above, these were classified as
"Pretenders" and were considered a third membership category.

Table 2:
RESPONSE BY GROUP

Radio
Television
Both
Total
Member
32
32
3
67
Non-Member
9
3
2
14
Pretenders
7
4
0
11
Total
48
39
5
92

The survey portion of the project used a four-wave mail technique.  The first
wave consisted of a personalized, hand-signed cover letter, a questionnaire, and a
postage-paid, return envelope.  The second, mailed a week later, consisted of a reminder
postcard.  The third wave, mailed two weeks later, was identical to the first wave (with a
modified cover letter), and was sent only to those stations that had not responded.  The
final wave was a second reminder postcard, mailed a week and a half after the third wave.
All data were input into the SPSS for Windows (v. 10.0) (Statistical Package for
the Social Sciences) computer program for analysis.  Hypotheses dealing with
membership were tested using one-way ANOVA; all other hypotheses were tested using
student t-test (one-tailed).  Because of the low response from non-members (with the
necessary division of non-members into two categories) and the exploratory nature of the
study, the decision rule was set at p<.10.

Results

        The first hypothesis, dealing with the relationship between economic factors and
membership, was partially supported.  A t-test of the differences between members and
non-members was not significant; an ANOVA test of the differences among all three
groups supports this hypothesis (members= 3.7, non-members=4.2, and pretenders=4.5;
F=3.1, p<.05).  Hypotheses 1b, dealing with the relationship between economic factors
and station type was not supported.
        Hypothesis 2a, predicting that members would rate peer pressure higher than non-
members was supported (non-members=3.1, members=3.8; t=2.4, p<.05).  No
relationship was found to support hypotheses 2b, relating station type to peer pressure.
        Hypothesis 3a was supported, with members rating the welfare of the industry
higher than non-members (non-members=12.1, members=13.1; t=1.2 p<.10).  The other
hypothesis relating station type with welfare of the industry was not supported.
        The final three hypotheses, dealing with the characteristics of pretenders, were
supported.  Hypothesis 4a predicted that pretenders would be more similar to non-
members than to members on rating importance of economic issues.  There was no
significant difference between pretenders and non-members on the economic dimension,
but pretenders (mean=3.7) were significantly different from members (mean=4.5) (t=2.2,
p<.05).  Hypothesis 4b predicted that pretenders would be more similar to members than
to non-members on rating importance of peer pressure.  No significant difference was
observed between pretenders and non-members, but pretenders (mean=3.1) were much
more likely than non-members to agree with the peer pressure item (mean=4.1) (t=2.4,
p<.05).  Finally, hypothesis 4c predicted that pretenders would be more similar to
members than to non-members on rating the importance of the welfare of the industry.
As expected, no significant difference was observed between pretenders and members on
this dimension, but pretenders (mean=13.3) scored significantly higher on the collective
good scale than non-members (mean=12.1) (t=1.2, p<.10).

Discussion

        The most important pattern in these results is the systematic relationship between
membership type and each of the dimensions of collective action identified.  These
results support the earlier findings of Meadows and Grant (1997), suggesting a rational,
economic basis for the existence of "free riders," as well as the more altruistic motives of
those interested in the public good..
        The identification of the "pretender" category of membership suggests further
consideration of Marwell and OliverŐs (1993) focus on the heterogeneity of resources
implicated in the collective action process.  These results suggest that the psychological
need of the pretenders to be involved (an "interest" resource) may be as important as the
economic resources in examining collective behavior.  This finding also supports
Marwell and OliverŐs (1993) identification of the production function as an underlying
tool for understanding collective behavior.  In this case, the decelerating production
function that characterizes membership in state broadcasting organizations creates the
opportunity for a station to be a pretender, feeling that they are part of the collective even
though they are free riders, taking advantage of the contributions of others to the
collective good.
        The presence of pretenders also underscores the dynamic nature of membership
organizations.  Old members may drop their membership at any point, and new members
are added regularly.  The result is that membership may always be in flux.  Incentives to
minimize membership churn may not only aid in retention but may also reduce the
number of pretenders attached to any organization.
        This dynamic nature of membership organizations also suggests consideration of
differences in TAB and its membership since the 1994 study.  The earlier study first
suggested the importance of collective action theory in understanding membership in the
organization.  As a result, TAB added a significant number of services designed to
benefit individual stations rather than the industry as a whole.  The result was a
substantial increase in membership in TAB since 1994.
        Another significant finding is the lack of difference between radio and television
stations on any of the three dimensions of collective action tested.  The reason for the
change from the 1994 results may be due to massive changes in the radio industry in the
intervening seven years.  The Telecommunications Act of 1996 required the FCC to
remove almost all ownership restrictions on radio stations, and the result has been a
massive consolidation in the radio industry.  The typical manager who ran two stations
(one AM and one FM) in 1994 is today running up to eight stations, with combined
revenues that are more similar to local television station revenues than to individual radio
station revenues.



Future Research

        As with the Meadows & Grant (1997) study, this project applied collective action
theory to one particular type of media industry trade organization, the state broadcast
organization.  Although use of the same organization as the focus of the study allows
greater comparability over time, similar studies are needed that deal with organizations
representing the interests of other media, as well as national and local organizations.
        The dynamic nature of membership in these organizations also suggests use of
case studies to explore specific reasons behind changes in membership status in these
organizations.  On an annual basis, most members renew their membership and most
non-members choose not to join, however pretenders are also established.  In-depth
analysis of those stations changing their membership status and the reason for the change
should illuminate underlying motivations, both related to collective action and to other
processes and motivations.

Conclusions

        Analysis of the media industry and audiences has long been focused on the
individual user or station, with little regard to interdependencies among the actors at each
level of analysis.  The application of collective action theory in this study is an important
step toward broadening our understanding of the processes that underlie the behavior of
media organizations and audience members.
        This study provides clear support for collective action theory, specifically for the
application of Marwell & OliverŐs concept of production functions.  A great deal more
research is needed to fully identify the range of  collective action processes underlying
media industry organizations.  That research can and should be directed with the
understanding of these organizations as examples of decelerating production functions,
and the related implications.

Bibliography
Goetze, David (1994). Comparing prisonerŐs dilemma, commons dilemma, and public
goods provision designs in laboratory experiments.  Journal of Conflict
Resolution, 38, 1, 56-86

Gold, Paul (1994). Two state broadcast associations and their constituents:  Money
talks?.  Feedback, 35, 2, 23-26.

Gould, Roger (1993). Collective action and network structure. American Sociological
Review, 58, 182-196.

Heckathorn, Douglas (1996). The dynamics and dilemmas of collective action.  American
Sociological Review.  61, 250-277.

Heckathorn, Douglas (1993). Collective action and group heterogeneity: voluntary
provision versus selective incentives. American Sociological Review, 58, 329-350.

Marwell, Gerald and Oliver, Pamela (1993). The critical mass in collective action: A
micro-social theory.  Cambridge, England:  Cambridge University Press.

Meadows, J. H. and Grant, A. E.  (1997).  Free-riders:  Problems with membership in
state broadcasting organizations   Paper presented at the 1997 Broadcast
Education Association Annual Meeting, April 1997, Las Vegas, NV.

National Association of Broadcasters (1993). State Association Survey.

Oliver, Pamela and Marwell, Gerald (1988).  The paradox of group size in collective
action:  A theory of the Critical Mass.  II.  American Sociological Review, 53, 1-8.

Oliver, Pamera, Marwell, Gerald, and Teixeira, Ruy (1985).  A theory of the critical
mass. I.   Interdependence, group heterogeneity, and the production of collective
action.  American Journal of Sociology, 91, 3, 522-556.

Olson, Mancur. (1965). The logic of collective action. Cambridge, MA:  Harvard
University Press.

Reisman, David (1990). Theories of collective action.  New York, NY: St. MartinŐs
Press.

Stoecker, Randy (1995). Community, movement, organization: The problem of identity
convergence in collective action.  The Sociological Quarterly, 36, 1, 111-130.

Taylor, Michael and Singleton, Sara (1993). The communal resource: Transaction costs
and the solution of collective action problems. Politics & Society, 21, 2, 195-214.

Texas Association of Broadcasters (1995). Annual Stakeholder Report.

Texas Association of Broadcasters (1991) Bylaws of the Texas Association of
Broadcasters.

Udehn, Lars. (1993). Twenty-five years with the logic of collective action. Ada
Sociologica, 36, 239-261.



                Free Riders and Pretenders 23